Click images for more details



Recent comments
Recent posts
Currently discussing

A few sites I've stumbled across recently....

Powered by Squarespace
« Why do good intentions in the public sector lead to evil? | Main | An also-ran in the climate prediction stakes »

Sinks and sources

Over the years I haven't really devoted much time to the carbon cycle, but I wonder if some of the attention of the climate debate will be switching to this area with the advent of the Orbiting Carbon Observatory, a satellite deployed last year to measure carbon dioxide concentrations across pretty much the whole planet.

The first maps were presented at AGU last month and the data is now publicly available:

Anthony has an interesting guest post from geophysicist Martin Hovland, which looks at some three hotspots that appear to be associated with subsea volcanic vents. I'm not sure whether these are of sufficient magnitude to provide backing for Ian Plimer's hypothesis about volcanic contributions to the carbon budget, but at least this represents a start in resolving the issues.

But the maps raise lots of other interesting questions too, at least for someone like me who is not particularly up to speed in this area. NASA seem to think there are no great surprises. Perhaps readers know some of the answers. For example, the low CO2 concentrations in the UK make no sense to me. And I wonder why India should have such low concentrations when China is so high. Why is there a hotspot in Greenland? And what is that one in the South Atlantic.

So many questions, so little time for reading.

PrintView Printer Friendly Version

Reader Comments (89)

the trick is that the image is at the lowest point of the annual curve. Have a look at the maximum in May.

Jan 2, 2015 at 11:15 PM | Unregistered CommenterHans Erren

Ruth Dixon
To me the fact that a tree continues to be a carbon sink throughout its life is not counter-intuitive. Most hardwoods (oak) in my experience put on mass on the outside and lose it internally, assuming one tree ring lost and one gained then two things mean that mass is gained in the trunk, first the diameter of the ring of material and secondly the height. In a mature tree 200+ years losing a 20 year ring involves considerably less mass than gaining a 200 year ring, to this non-expert of matters arboreal that will be the major contributor to mass gain or loss. Even when the tree dies not all carbon is returned to the atmosphere as CO2 so the life of a tree is an overall carbon sink

I think that overall atmospheric CO2 content has declined since the evolution and spread of trees.

Jan 2, 2015 at 11:36 PM | Unregistered CommenterSandyS


I think that overall atmospheric CO2 content has declined since the evolution and spread of trees.

Yes indeed, it had an enormous effect - see for example Figure 5 in this paper, which shows the massive drop in CO2 during the Devonian period (about 420 to 360 million years ago) due to the rise of trees, which hugely increased removal of CO2 from the atmosphere by weathering.

Jan 3, 2015 at 12:07 AM | Registered CommenterRichard Betts

Richard, w.r.t the Amazon and India I was wondering about the graph in the post above, i.e. OCO.

It is not plausible that deforestation-mediated carbon loss contributed to a local rise in atmospheric CO2 as seen in the data. The South American CO2 cloud is mostly directly over the tropical evergreen forest. We know it doesn't burn as much as savannah, and nor can clear-cutting put CO2 directly into the atmosphere without intervening wood/leaf decomposition. Though there is logging in the Amazon, it is obviously not diffusely present throughout the entire Amazon.

Jan 3, 2015 at 12:27 AM | Registered Commentershub

Oh, no, not >i>another Carbon Cycle 101 fail!

On the undersea side , his Eminence may not be " sure whether these are of sufficient magnitude to provide backing for Ian Plimer's hypothesis about volcanic contributions to the carbon budget, but .." no geochemist of my acquantance shares his doubts-- > 100 times more CO2 flows from people buring fossil fuels than volcanic vents wet and dry, andHhenry's Law makes it a forgone conclusion that undersea vents contribute less to the atmospheric burden than vulcanism on land.

Jan 3, 2015 at 12:49 AM | Unregistered CommenterRussell

Michael hart

Not totally sure why the poles aren't covered. The satellite measures land and ocean in sunlight and to angles of max 85 degrees (nadir mode - from documentation). Maybe it just hasn't mapped enough data yet.

Jan 3, 2015 at 12:59 AM | Unregistered CommenterMicky H Corbett

"Michael Hart, Jaime Jessop and others, it's important to realise that (as some have pointed out) the CO2 map is only a few weeks of data. Importantly, it's at the time of year when northern hemisphere CO2 is at it's lowest (see the Mauna Loa data). Through the NH summer, growing vegetation takes up CO2 so atmospheric concentrations are reduced - from November onwards,..."

Yes indeed, Richard. I have pointed out the same myself and acknowledged that it is preliminary data. But the mixing time between the Northern and Southern hemispheres (from, say, the d14C bombspike curves) is clearly great enough for us to examine the behaviour of distinct regions in different hemispheres with this data.

And other points I make are not dependent on whether it is Northern Hemisphere Summer or Winter. With only minor seasonal variations, human fossil fuel emissions proceed 24/7/365. Your point about it being a total-column-depth is a good one. But I think human contributions should be visible all the time if the instrument is giving a true reading and the mantra of CO2 being a well mixed gas are both true. I think one of them is not. Perhaps we can discuss that some time.

Jan 3, 2015 at 1:04 AM | Unregistered Commentermichael hart

Brazil is in the middle of a record drought that has extended through the 2013-14 rainy seasons, the 2014 dry season and the onset of the rainy season in 2014 has been delayed into December. A delayed rainy season is typical of El Nino conditions. While a full El Nino has not been declared, it is really kind of sortta el Ninoish out there.

The fire maps very much overlap the CO2 column depth maps

While we do not have data yet from 2014 for CO and aerosols, in previous years, the ag burning showed that CO and aerosols overlapped the CO2 when ag burning was the source of the CO2. You can see this in AIRS maps of CO during the burning season

Jan 3, 2015 at 1:12 AM | Unregistered CommenterEli Rabett

That is, I think at least one of them is not valid for the points under discussion.

Jan 3, 2015 at 1:13 AM | Unregistered Commentermichael hart

This CO2 map is certainly much more informative than a single trace from Moana Loa. But what about the 3rd dimension? How is the CO2 distributed vertically in the atmosphere? Is it concentrated near the surface - or is the concentration more-or-less constant for the full-height?

Does anyone know if vertical data is available?

* * * *
We also need to always bear in mind that there is a carbon *cycle* at work. The natural sources (including some level of volcanism and perhaps much of the scrub-burning) may be very large, but they are theoretically balanced by equally large carbon sinks - forests, other biomass, oceans, corals and limestone deposition, etc.

Going back to Global-warming 101, the big spanner in the works is the CO2 from burning fossil fuels - and the idea that this is tipping the balance of the Carbon Cycle. It's basic but some of the above comments seem to forget this.

Jan 3, 2015 at 3:32 AM | Unregistered CommenterJ Calvert N

People are reading this map from different points of view.

Some areas might have elevated CO2 that result from decay of organic matter, emissions associated with tectonic plates etc, and burning of organic matter and fossil fuels.

But the big latitudinal belt near the equator more or less correlates with the Inter-Tropical Convergence Zone.(ITCZ) and results at least in part from out-gassing of CO2 from the oceans.

Jan 3, 2015 at 7:26 AM | Unregistered CommenterFrederick Colbourne

Eli Rabett

So Brazil is in the middle of a record drought ,okay so in what century did the Brazilians decide to start keeping records ?

Jan 3, 2015 at 8:25 AM | Unregistered Commenterjamspid

Richard, you say,

"Michael Hart, Jaime Jessop and others, it's important to realise that (as some have pointed out) the CO2 map is only a few weeks of data. Importantly, it's at the time of year when northern hemisphere CO2 is at it's lowest (see the Mauna Loa data). Through the NH summer, growing vegetation takes up CO2 so atmospheric concentrations are reduced - from November onwards, CO2 starts to rise again after the fallen leaves of autumn decay and release CO2 back to the atmosphere. "

The NASA graphic states: "Global atmospheric carbon dioxide concentrations from Oct. 1 through Nov. 11, as recorded by NASA's Orbiting Carbon Observatory-2."

So we're looking at Autumn/Fall/early winter really, as CO2 is on the rise. Compare this to NASA's modeled CO2 distribution: (Oct 4th) (Oct 23rd) (Nov 10th)
Entire video here:

In actual fact, considering the real-time data graphic represents an average over this period, the deviation from modeled and actual data is not too different for the northern mid latitudes to the equatorial regions, but the divergence becomes increasingly obvious as we progress on into the NH winter. Which is why it will be very instructive to see the data from Nov 1st onwards and why it is rather a shame that there appears to be no coverage of the far northern latitudes.

Jan 3, 2015 at 9:58 AM | Unregistered CommenterJaime Jessop

Nov 11th onwards, I should have said.

Jan 3, 2015 at 10:10 AM | Unregistered CommenterJaime Jessop

What this image tells us are momentary concentrations of CO2. It is about as useful as a temperature map of the globe in absolute values, showing that it's hot on equator and cold on poles.

Best thing to do with this picture is to compare it to the NASA's model output. Particularly, this picture comes from October, so make sure to move the slider to October in the video, too. Then you'll see there indeed are no (big) surprises.

Surprises may eventually come from differences between the model and measured data. Only these will tell use where there are sources and sinks we don't know about (or have estimated incorrectly). But these differences will be order of magnitude smaller than the difference between minima and maxima in this image.

Jan 3, 2015 at 10:30 AM | Unregistered CommenterKasuha

Don't be afraid to say "I don't know"
- It's one thing to claim to be able to interpret super complex data.
- It's another thing to be able to spot errors in that interpretation.
- As ever I make no claim to have climate clairvoyance, but will continue to try to spot flaws in the reasoning of those that do.
- "Those skeptics they never offer a better model than the IPCCs" is a common climate crusader fallacy.

Jan 3, 2015 at 11:05 AM | Registered Commenterstewgreen

I don't know a great deal about this project but I do find myself left with some questions:

1. What was the purpose of this investment in the OCO?

2. There clearly was an expectation of data and I suspect a political timing of first release, so what was the expectation?

3. How did they ever expect to see a 3.2% signature across the globe of anthroprogenic contribution to total world CO2 emissions in a chaotic climatic system?

I think the answer to 2 & 3 clearly amplify the answer to 1. I would not be surprised if this project will be dead in a year, the data didn't justify the investment and it seems bizarre they ever thought it would be the golden chalice to justify the models. All of which is a shame, as observational science is so important. I would be very to be enlightened if anyone so wishes to do so. I try to keep an open mind but confess to a growing cynicism where I am questioning motives more than the "data"

Jan 3, 2015 at 12:49 PM | Unregistered CommenterBadgerbod

I would not be surprised if this project will be dead in a year, the data didn't justify the investment and it seems bizarre they ever thought it would be the golden chalice to justify the models.

Well at least they didn't spend the money on more models, or worse, reaching out to Muslims. [not that I have anything against Muslims, well the vast majority of them at least].

Jan 3, 2015 at 2:18 PM | Registered Commenterlapogus

As this system has completed 2 complete scans of the earth: Oct 1-16 and Nov 2-17, I expect it will be to able to report slow changing levels of CO2. As each area of the surface has had 2 measurements taken in this time, I think we will need to wait a rather long time before the data can report anything significant.

Jan 3, 2015 at 2:22 PM | Unregistered CommenterSteve Richards

As for why there are no data from the Arctic there is no mystery. OCO works by spectrometry of reflected sunlight in the near infrared, so there will never be any data from the Arctic or Antarctic in winter. Not enough sunlight.

As for Eli Rabbet’s claim of “no surprises”, I don’t agree. The pattern shown is remarkably different from earlier measurements at the same season (including those ER links to, I wonder if he actually looked at them). About the only thing similar are high concentrations over parts of Brazil and Central Africa. These are probably due to biomass burning (though not in Central and Western Amazon – too wet for burning).

Some patterns are clear. For example oceanic upwelling areas tend to have high concentrations – very likely due to oceanic outgassing. However, in many cases the pattern is quite mysterious. For example the high concentration over Denmark Sound between Iceland and Greenland. Of course it has nothing to do with icelandic volcanism – it’s upwind from Iceland - nor with US fossil fuel burning, unless the CO2 is moved by pipeline to Greenland. It might possibly be caused by the very cold waters of the East Greenland Current meeting the Gulfstream and outgassing, particularly in winter when there is very little phytoplankton activity.
Also the very low level over India is incomprehensible. As everyone who has ever visited India knows, biomass burning is very intensive there, and by October-November it is dry and fairly cool so plant uptake of CO2 is down. Somebody very ingeniously suggested above that the biomass burning does not show up since it mostly happens indoor, but since Indians do not normally live in gastight houses that is not a viable hypothesis.

Other mysteries: why higher levels in Labrador-Ungava (forested, frozen and uninhabited) than in New England (which is none of these things).
And why the big concentrations off the western Aleuts and N of New Zealand? In these two cases a tectonic explanation does seem possible.

Jan 3, 2015 at 5:16 PM | Unregistered Commentertty

Reading tty's analysis gives me the suggestive hope that given this new vision, some genius will figure this out, like Tisdale with the oceans.

Jan 3, 2015 at 10:48 PM | Unregistered Commenterkim

Other mysteries: why higher levels in Labrador-Ungava (forested, frozen and uninhabited) than in New England (which is none of these things).

You are assuming that during summer there are no growing plants in Labrador and that they are not dying off and decaying during the early/mid Fall.

Jan 3, 2015 at 11:11 PM | Unregistered CommenterEli Rabett

tty, you are right. I put the question to Betts: why is there no increased CO2 concentrations over the Indian subcontinent - there is open burning, indoor fuel consumption and industrial activity during the period of measurement in the OCO. Betts answered that cook fires are indoors and MODIS doesn't capture them. This was particularly a #facepalm moment - MODIS doesn't do cookfires, I can understand that!! It has difficulty seeing forest fires themselves, below a certain size. Neither head nor tail could be made of Betts' response, I let it pass as one written somewhat hastily and with satellites mixed.

eli rabett, you do realize you are not refuting tty's point, only supporting it? If there is woodland and crop burning from MODIS data over India, why is no noticeable CO2 cloud being registered?

Jan 4, 2015 at 4:17 AM | Registered Commentershub

"You are assuming that during summer there are no growing plants in Labrador and that they are not dying off and decaying during the early/mid Fall."

October-November is not "early-mid fall" in the subarctic. The whole area was snow-covered by the end of October (as usual):

"The problem, if Eli may venture a guess, is that many make a guess and then assume it to be true rather than checking”

The same to you Br'er Rabbit. And some advice: idiot snarkery is not a very effective debating gambit.

Jan 4, 2015 at 7:27 PM | Unregistered Commentertty

Max temperatures at Gander Labrador were above 10 C for October 2014 and min temperatures were above freezing. It cooled in November but still for most of the month temperatures were above freezing. There were 293 heat degree days and 0 cold degree days. In November there were 484 hot degree days and no cold degree days.

Hot times in Labrador this year

Jan 4, 2015 at 8:49 PM | Unregistered CommenterEli Rabett

Sorry about that, that was Gander in Newfoundland, but here is Goose Bay in October. Temperatures above freezing, 388 hot degree days no cold ones. November was also warm, with 726 hot degree days and only turned cold at the end of the month.

Pretty much all of Labrador and New Foundland was warm this October and November and had no snow, at least where the weather stations were

Jan 4, 2015 at 9:25 PM | Unregistered CommenterEli Rabett

A bit late to this… erm... fire... but I’ll press on, anyway: one curious thing is that people equate fire with human activity. Yet there are many eco-systems that are dependent on fire: regions of South Africa, for example, or the southern United States – or vast areas of Australia. Fire, it seems, has been a lot more prevalent than humans, in times past. Why do we now think that plants that have evolved such that they are dependent on fire should be protected from it? Perhaps it is because we have learned how to control fire, and in doing so, have given ourselves far greater credit, kudos or esteem than we are really worth. Fire can, and does, occur quite naturally.

And if we burn the fire indoors, will that make a difference to the CO2 outside? Well, unless everybody is burning it in order to asphyxiate themselves, yes, it will. Bit of a flaw in your argument, there, Mr Betts.

Jan 4, 2015 at 9:39 PM | Unregistered CommenterRadical Rodent

Dennisa, thanks very much for your comment.

There is an element of mysticism in Western views about the Amazon which flies in the face of facts. And this is a good time to remind those who have not read it already to check out "Carry On Up the Amazon" at Geoff Chambers' blog, one of the funniest blog posts ever. It concerns a visit by Vivienne Westwood and her acolytes to the Amazon, where among other indignities their underpants were eaten by ants.

Remember a few years ago when we were being told that a terrible drought in the Amazon was going to wipe out the rainforest? All due to AGW?

We don't hear much about it these days - another failed prediction gone to the forgettory.

Jan 4, 2015 at 10:54 PM | Registered Commenterjohanna

johanna, check out this interesting paper titled 'Wash and Spin Cycle Threats to Tropical Biodiversity' by authors Rhett Butler, William Laurance among others.

Koh, L. P., Ghazoul, J., Butler, R. A., Laurance, W. F., Sodhi, N. S., Mateo-Vega, J. and Bradshaw, C. J. A. (2010), Wash and Spin Cycle Threats to Tropical Biodiversity. Biotropica, 42: 67–71.

The URL is here:

Jan 4, 2015 at 11:19 PM | Registered Commentershub

Thanks, shub, but I could only read part of the abstract due to paywalling.
Have you read the whole thing? What does it say?

Jan 5, 2015 at 12:08 AM | Registered Commenterjohanna

johanna --
Non-paywalled version is available here.

Jan 5, 2015 at 2:04 AM | Registered CommenterHaroldW

johanna, if you follow the Amazon literature, it becomes evident there are cycles of 'concern' coincident with the prevalent decadal scale climatic trend (with a baseline of *all* authors assuming as an unstated given that having more trees in the Amazon is a 'good thing'). In the late 90s, there was a spurt of studies into Amazon fire and fire risk - coinciding with the El Nino of '98. In the next decade, came the drought-will-destroy-Amazon studies prompted by a couple of droughts. In other words, there are definitely cycles of academic 'blackwash' - the concept set forth in the paper, though the paper describes environmental activists practicing it.

Jan 5, 2015 at 3:07 AM | Registered Commentershub

Thanks, Harold and shub. It is a reasonably balanced paper, although still green-tinged.It's streets ahead of the usual mindless Gaia-worship, though.

Jan 5, 2015 at 9:42 AM | Registered Commenterjohanna

My position on it is this - it's more data. If it can be used as is for their(warminsts) purpose, it will. If it is diametrically against the warmist position, it will be overlooked, just like all other data, until someone tries to make use of it. Then it will be adjusted to make it tell the correct story.

It should yield information useful to scientists, but there aren't many of them around anymore since most of them seem to have lost their sense of integrity while chasing grant money.

Jan 5, 2015 at 6:38 PM | Unregistered CommenterTom O

Eli Rabbett: You just never give up however wrong you are, eh? Goose Bay is a coastal station, and so considerably warmer than most of Labrador. Even so first snow arrived on Nov 2 and average snow depth in November was about 30 cm. Average temperature for the month was -6.2 degrees C.
Unfortunately there are practically no inland weather stations in Labrador, and the few there are apparently don't record snow depth, so I'm afraid we will have to be content with the satellite data. By the way I know from personal experience that they are quite reliable in discerning snow-cover, at lerast in Europe, so I don't think they can be that unreliable in North America.

Once again here is the image that shows that the area we are discussing became snow-covered the last days in October:

And here is departure from normal + climatology that shows that this is a completely normal date, or perhaps slightly earlier than normal:

Incidentally, if you aren't aware of it, temperatures below freezing is a necessary but not sufficient precondition for long-term snow-cover.

Jan 5, 2015 at 6:57 PM | Unregistered Commentertty

LATE in October. However, if you had looked at the map for where in Labrador there are weather stations you would see that just about all of them are on the coast with the exception of Churchill Falls. And if you looked at the weather reports for Churchill Falls, you would see no snow in October and temperatures above freezing. Anybunny interested can check this through the link by clicking on the stations and then going down the current weather report page for historical weather. Check any of the stations (looks like Labrador City has no historical data for 2014 tho.)

The first report for snow on the ground from Nain, the northern most station is November 5. The first report of snow on the ground from Goose Bay is Nov 2 although there were days towards the end of October when it snowed, but from the wether reports it did not appear to stick because there is no report of snow on the ground

Jan 8, 2015 at 1:55 AM | Unregistered CommenterEli Rabett

National Snow and Ice Data Centers

October 1:

October 15

October 26

Jan 8, 2015 at 6:08 AM | Unregistered CommenterEli Rabett

"our knowledge is pretty limited.", at least you have multiple persons. Or were you projecting YOUR ignorance onto everyone? Well, ha, ha.

Apr 2, 2015 at 12:25 PM | Unregistered CommentercRR Kampen

PostPost a New Comment

Enter your information below to add a new comment.

My response is on my own website »
Author Email (optional):
Author URL (optional):
Some HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <em> <i> <strike> <strong>